LightPath Technologies Inc (LPTH) 2026 Q4 法說會逐字稿

Good afternoon, ladies and gentlemen. Thank you for standing by. Welcome to LightPath Technologies' fiscal fourth-quarter and full-year 2026 earnings conference call. (Operator Instructions) This conference is being recorded today, September 10, 2026, and the earnings press release accompanying this conference call was issued after the market closed today.

各位女士、先生，下午好。感謝各位耐心等候。歡迎參加 LightPath Technologies 2026 會計年度第四季及全年財報電話會議。(接線員指示) 本次會議於今日（2026 年 9 月 10 日）錄音，且本次電話會議所搭配的財報新聞稿已於今日收盤後發布。

I'd like to remind you that during the course of this conference call, the company will be making a number of forward-looking statements that are based on current expectations, involve various risks and uncertainties, and discussed in its periodic SEC filings. Although the company believes that the assumptions underlying these statements are reasonable, any of them could be proven to be inaccurate, and there can be no assurances that the projected results will be realized.

提醒各位，在本次電話會議期間，公司將發表若干前瞻性陳述，該等陳述係基於目前預期，涉及各種風險與不確定性，並已於公司定期向美國證券交易委員會（SEC）提交的文件中討論。儘管公司相信支撐這些陳述的假設屬合理，但其中任何假設都可能被證明不正確，且無法保證預測結果一定能夠實現。

In addition, references may be made to certain financial measures that are not in accordance with generally accepted accounting principles or GAAP. We refer to these as non-GAAP financial measures. Please refer to our SEC reports in certain areas of our press releases, which include reconciliations of non-GAAP financial measures and associated disclaimers.

此外，會議中可能會提及某些不符合一般公認會計原則（GAAP）的財務衡量指標。我們將其稱為非 GAAP 財務衡量指標。請參閱我們的 SEC 報告以及新聞稿中的特定章節，其中包含非 GAAP 財務衡量指標的調節表及相關免責聲明。

CEO, Sam Rubin, will begin today's call with a strategic overview of the business and recent developments for the company; while CFO, Albert Miranda, will then review financial results for the quarter and the fiscal year. Following the prepared remarks, there will be a formal question-and-answer session.

執行長 Sam Rubin 將以公司策略概覽及近期發展作為今日電話會議的開場；接著財務長 Albert Miranda 將回顧本季及本會計年度的財務結果。在預先準備的發言之後，將進行正式的問答環節。

I'd like to now turn the conference over to CEO, Sam Rubin. Sam, the floor is yours.

現在我想把電話會議交給執行長 Sam Rubin。Sam，請開始。

Thank you, operator. Good afternoon to everyone, and welcome to LightPath Technologies' fiscal fourth-quarter and full-year 2026 financial results conference call.

謝謝接線員。各位下午好，歡迎參加 LightPath Technologies 2026 會計年度第四季及全年財務結果電話會議。

The last few calls, I typically opened by talking about the strategy and how the strategy is working and where it's taking us. Tonight instead, I will let the numbers talk and let the fiscal year results do the talking.

在過去幾次電話會議中，我通常會先談策略、策略如何發揮作用，以及它將帶領我們走向何處。但今晚我將改為讓數字說話，讓本會計年度的結果來說話。

Fiscal 2026 is the first year in which the transformation we have been describing shows up cleanly in every line of the financial statements, not just the backlog. Four numbers frames a year.

2026 會計年度是我們一直在描述的轉型首次在財務報表的每一個項目上都清楚呈現的一年，而不僅僅是在訂單積壓（backlog）上。四個數字勾勒出一整年。

Revenue grew 93% from $37 million to nearly $72 million. Gross margin expanded from 27% to 36%. Adjusted EBITDA moved from a $5.1 million loss to a $4.2 million profit, a swing of more than $9 million. And backlog finished at $110.9 million, up 197% from where we started the year. The fourth quarter was our best quarter in every one of those categories.

營收從 3,700 萬美元成長 93%，接近 7,200 萬美元。毛利率從 27% 擴大至 36%。調整後 EBITDA 從虧損 510 萬美元轉為獲利 420 萬美元，改善幅度超過 900 萬美元。訂單積壓期末為 1.109 億美元，較年初成長 197%。第四季在上述每一項指標上都是我們表現最佳的一季。

Revenue of $21.2 million was a company record and our fourth consecutive quarter of sequential growth. Gross margin was 39.4%. Adjusted EBITDA was $2.1 million, or 10% of revenue, which was our fourth straight profitable quarter on that measure. All great results, which we expect will continue to grow and improve.

營收 2,120 萬美元創公司紀錄，且為連續第四季季增。毛利率為 39.4%。調整後 EBITDA 為 210 萬美元，約占營收 10%，這也是我們在該指標下連續第四季獲利。這些都是非常出色的成果，我們預期將持續成長並進一步改善。

Now, I'd like to spend a moment on the quality of that margin, because it is the part that I am most pleased with. The 39.4% margin did not come from a one-time favorable contract or from raising prices. It did come from two things we did.

現在我想花一點時間談談這個毛利率的「品質」，因為這是我最滿意的部分。39.4% 的毛利率並非來自一次性的有利合約，也不是因為調漲價格。它來自我們做的兩件事。

First is the mix of products. Assemblies, modules, and cameras were 43% of the fourth quarter revenue and 44% of the full year, compared to 23% of the revenue in fiscal 2025. These products, which our strategy has really took us towards, have both higher prices and higher margins as a result of the significant value add compared to our legacy component business.

第一是產品組合。組裝件、模組與相機占第四季營收的 43%，占全年營收的 44%，相較於 2025 會計年度僅占 23%。這些產品正是我們策略真正引導我們前進的方向；由於相較於我們傳統的元件業務具有顯著的附加價值，因此其售價更高、毛利率也更高。

Second is execution. The yield and throughput problems that dragged our component margins in the past are, well, they're in the past. And every one of our four product groups improved its margin year over year. The mix improvement is a result of strategy.

第二是執行力。過去拖累我們元件毛利率的良率與產出（throughput）問題，嗯，已經成為過去式。而且我們四大產品群的毛利率皆較去年同期提升。產品組合的改善是策略的成果。

The margin improvement is a result of operation. We needed both, and this year we got both. And while the backlog has not grown sequentially, shortly after we closed the quarter, we reported two large orders totaling $24 million and have been continuing to book and grow our backlog. So the backlog you will see for the quarter ending in about three weeks' time will show already some more growth.

毛利率的提升是營運的成果。兩者缺一不可，而今年我們兩者都做到了。此外，雖然訂單積壓並未呈現季增，但在季度結束後不久，我們公告了兩筆合計 2,400 萬美元的大型訂單，並持續接單、推升訂單積壓。因此，各位在大約三週後結束的那一季所看到的訂單積壓，將已反映出進一步的成長。

While we continue securing new customers, both by converting them to BlackDiamond and by providing them with assemblies and systems, much of the orders coming now, like the two large orders I just mentioned from July, are for production as many of the programs we have been working on move from qualification to production.

在我們持續取得新客戶的同時——包括將其轉換至 BlackDiamond，以及向其提供組裝件與系統——目前進來的許多訂單（例如我剛提到的 7 月兩筆大型訂單）是用於量產，因為我們一直推進的多個專案正從認證（qualification）階段轉入量產階段。

We have said in the past that a design and qualification of a new program, whether it's a redesign to use BlackDiamond or a new program altogether, can take up to two years. Much of the growth in our backlog and bookings recently that we are seeing is the result of such new programs beginning to move into production.

我們過去曾表示，一個新專案的設計與認證——無論是為了改用 BlackDiamond 的重新設計，或是全新的專案——可能需要長達兩年。我們近期看到的訂單積壓與接單量的許多成長，正是源自這類新專案開始轉入量產。

Two years ago, in late 2024, China imposed the first restrictions on export of germanium and gallium, and we began to see a growth in demand to BlackDiamond and systems using BlackDiamond glass. Now we are beginning to see the transition of some of those into production.

兩年前，也就是 2024 年底，中國首次對鍺與鎵的出口施加限制，我們開始看到對 BlackDiamond 以及使用 BlackDiamond 玻璃之系統的需求成長。現在我們開始看到其中部分需求正轉入量產。

A trend I expect will intensify, as most customers did not start their redesign and substitution effort until much after the initial export ban on germanium. To that end, I will provide now an update on some of our key programs we have in the pipeline and their status.

我預期這一趨勢將會加速，因為多數客戶直到鍺的初次出口禁令後很久才開始進行重新設計與替代的工作。為此，我現在將就我們管線中的一些關鍵專案及其進度提供最新資訊。

As mentioned in recent calls, due to the good problem of experiencing exponential growth in all our sales fronts, I can't really cover all the large programs. So instead, I will focus on ones where we had some changes or recent developments. NGSRI, our three-year-old interceptor program with Lockheed Martin.

如同近期電話會議所提到的，由於我們在所有銷售面向都面臨指數型成長這個「甜蜜的煩惱」，我無法逐一涵蓋所有大型專案。因此，我將聚焦在近期有變化或最新進展的專案。NGSRI——我們與洛克希德馬丁（Lockheed Martin）合作、已進行三年的攔截器專案。

As many have heard, the Army has pushed out the timeline by a few months in an effort to potentially explore other options. We do not see this as a risk to us, only a delay. We have seen this happen in multiple other programs where the Army wants to foster a truly competitive environment. Our confidence continues to be very high. And given that our seeker is being designed and evaluated into multiple platforms now, we have little concern here.

如同許多人所聽到的，陸軍將時程往後推遲了幾個月，以便可能探索其他選項。我們不認為這對我們構成風險，只是延後。我們在其他多個專案中也見過類似情況，陸軍希望營造真正具競爭性的環境。我們的信心仍然非常高。而且鑑於我們的導引頭（seeker）目前正被設計並評估導入多個平台，我們對此幾乎沒有疑慮。

In the last few months, we have relocated the groups that works on that seeker into a new building and have begun investing in increasing capacity for building seekers, knowing that any program that moves into production will need to scale very quickly.

在過去幾個月裡，我們已將負責該尋標器的團隊搬遷至一棟新大樓，並開始投資提升尋標器的製造產能，因為我們知道任何進入量產的專案都需要非常快速地擴產。

More broadly about seekers and missiles. Our camera systems are now being designed into or actively evaluated in seven different platforms, three of which are with Lockheed Martin. The remainder are with primes, or as they're sometimes referred to recently, neo-primes, newer companies entering the defense market. The full qualification of our low-cost seeker that was completed as a result of the NGSRI flight test, as expected, opened the door to many other opportunities.

更廣泛地談尋標器與飛彈。我們的相機系統目前正被設計導入或正積極評估於七個不同的平台，其中三個與洛克希德·馬丁（Lockheed Martin）合作。其餘則與主承包商合作，或如近期有時所稱的「新主承包商」（neo-primes），也就是進入國防市場的較新公司。我們低成本尋標器的完整鑑定已因 NGSRI 飛行測試而完成，且如預期般，為許多其他機會打開了大門。

The same manufacturing facility in Texas will support all of those opportunities. In border patrol or border tower, we have seen funding being released from DHS to the primes. However, that has not yet translated into orders for cameras. What we have seen is a growth in demand for similar towers and cameras that end up installed outside the US, primarily in the Middle East.

德州的同一座製造設施將支援所有這些機會。在邊境巡防或邊境塔方面，我們看到國土安全部（DHS）已向主承包商釋出資金。然而，這尚未轉化為相機訂單。我們看到的是，對類似塔台與相機的需求正在成長，而這些最終安裝地點在美國境外，主要在中東。

A few other programs. We have an unnamed airborne program which has completed qualification and we await the production order for LRIP. The Apache program is looking like it might make a comeback soon with a renewed interest in that system. Drones, and in particular drone dominance programs, are generating significant demand, which we are addressing by starting to add automation to some of our processes for high-volume assemblies.

另外幾個專案。我們有一個未具名的空載專案已完成鑑定，我們正等待 LRIP 的量產訂單。阿帕契（Apache）專案看起來可能很快會回歸，因為對該系統的興趣再度升溫。無人機，尤其是無人機制霸（drone dominance）相關專案，正在帶來顯著需求；我們正透過在部分流程導入自動化來因應高產量組裝的需求。

And counter-UAS programs continue to move along well, with two of the programs now transitioning to a cadence of deliveries of tens of systems a month. In parallel to more design wins of our existing products, the teams have been working on designs and redesigns of additional products, all of which leverage our BlackDiamond glass and make use of our supply chain resilience and having alternative materials instead of depending on germanium.

反無人機（counter-UAS）專案也持續順利推進，其中兩個專案目前正轉入每月交付數十套系統的交付節奏。在既有產品持續取得更多設計導入（design wins）的同時，團隊也在進行其他產品的設計與再設計；這些產品皆運用我們的 BlackDiamond 玻璃，並利用我們供應鏈韌性與替代材料，以降低對鍺（germanium）的依賴。

On the camera front, we have been redesigning the last of the G5 cooled cameras to use BlackDiamond instead of germanium. That program is progressing well technically, though behind schedule. However, all the results we have seen so far indicated the cameras will work at least as well as the germanium-based cameras.

在相機方面，我們一直在重新設計最後一款 G5 冷卻式相機，將材料由鍺改為 BlackDiamond。該專案在技術上進展良好，但進度落後。不過，截至目前我們看到的所有結果都顯示，這些相機的表現至少會與以鍺為基礎的相機同等。

We have also been working on zoom lenses and zoom cameras in what is called long-wave infrared, often referred to as uncooled cameras. Here, too, we identified an area of the market in which we can leverage our position to provide products without supply chain constraints. An effort that has been ongoing for the last year is now coming to fruition, with our first orders for zoom lenses and complete uncooled zoom cameras to be delivered soon. Once those are fully production ready, we expect to see the need in the market translate into orders fairly quickly.

我們也一直在研發所謂長波紅外（long-wave infrared）的變焦鏡頭與變焦相機，這類產品常被稱為非冷卻式相機（uncooled cameras）。在這裡，我們同樣找到了可利用自身優勢、提供不受供應鏈限制之產品的市場區塊。一項已持續一年的工作如今正開花結果，我們的首批變焦鏡頭與完整的非冷卻式變焦相機訂單將於近期交付。一旦這些產品完全具備量產條件，我們預期市場需求將相當快速地轉化為訂單。

Let me now turn to the structural changes we have completed after year-end, which is the divestiture of our China operation. In July, we signed a definitive agreement to sell our China subsidiary to an entity owned by the local management team that has run the facility for us for $4.5 million, payable in installments over five years. And that transaction is expected to close later this month.

接下來我談談我們在年末後完成的結構性調整，也就是出售我們在中國的營運。7 月，我們簽署了最終協議，將中國子公司出售給由當地管理團隊持有的實體；該團隊一直為我們營運該設施，交易金額為 450 萬美元，將於五年內分期支付。該交易預計將於本月稍晚完成交割。

Financially, it is a modest event. Roughly $4.5 million of annual third-party revenue leaves our consolidated results. And the buyer continues to supply us as a third-party vendor for our commercial customers in the US and Europe. So there is no disruption to those customers.

就財務面而言，這是一件影響不大的事件。約 450 萬美元的年度第三方營收將自我們的合併報表中移除。而買方仍將以第三方供應商身分，為我們在美國與歐洲的商業客戶供貨。因此不會對這些客戶造成中斷。

Strategically, it is not modest at all. Six years ago, most of our manufacturing footprint was in China, and more than 1/3 of our revenue came from China. As of this fall, LightPath now does everything: melts glass, coats optics, builds cameras and assembly exclusively outside of China, in Orlando, in Plano, Texas, in Hudson, New Hampshire, and in Riga, Latvia.

但在策略面上，這一點也不小。六年前，我們大部分的製造據點都在中國，且超過 1/3 的營收來自中國。截至今年秋季，LightPath 現在所有作業都在中國以外完成：玻璃熔融、光學鍍膜、相機製造與組裝，據點包括奧蘭多、德州普萊諾（Plano, Texas）、新罕布夏州哈德遜（Hudson, New Hampshire）以及拉脫維亞里加（Riga, Latvia）。

We have no ownership, no facility, and no commercial activity in China. For a customer base that is now dominated by the defense primes and public safety agencies, that is no longer a nice-to-have talking point, but something that will now open up significant bid opportunities where that is a condition for participation. That connects directly to our regulatory backdrop.

我們在中國沒有任何持股、沒有設施，也沒有任何商業活動。對於目前以國防主承包商與公共安全機構為主的客戶群而言，這不再只是「加分的說法」，而是將為我們打開大量投標機會的關鍵條件，因為這已成為參與投標的必要門檻。這也直接連結到我們的監管背景。

Defense programs are required to move off optical glass and optical components sourced from covered nations before the end of the decade. What has changed over the last year is not the rule, it is the timing. Qualification cycles for an optical system run two to three years. So the sourcing decisions that determine who supplies those programs in 2029, 2030, are being made now in this fiscal year and in the next one.

國防專案被要求在本十年結束前，停止使用來自受涵蓋國家（covered nations）的光學玻璃與光學元件。過去一年改變的不是規則，而是時程。一套光學系統的鑑定週期為兩到三年。因此，決定 2029、2030 年由誰供應這些專案的採購來源決策，正在本會計年度與下一個會計年度做出。

However, several executive orders around waivers for germanium and classification of material as critical and specific tariffs applied would likely pull that timeline even closer.

然而，若干與鍺豁免（waivers）、將材料分類為關鍵（critical）以及適用特定關稅等相關的行政命令，可能會使該時程進一步提前。

Now, to address that, let me talk about capacity a bit, because capacity is our single biggest operational theme going into fiscal 2027. When we acquired Amorphous Materials in January, we increased our BlackDiamond melting capacity. And just as importantly, we unlocked large diameter melting, up to 10 inches and beyond, versus the five inches we could produce before. In optics, the further you need to see, the larger the optics need to be.

為了因應這點，我想談談產能，因為在邁向 2027 會計年度時，產能是我們最重要的單一營運主題。我們在 1 月收購 Amorphous Materials 後，提高了 BlackDiamond 的熔融產能。同樣重要的是，我們解鎖了大口徑熔融能力，可達 10 吋甚至更大，相較之下我們先前只能生產 5 吋。在光學領域，你需要看得越遠，就需要越大的光學元件。

Large diameter is what opens up long-range camera systems, large assemblies, and space-based missile detection and tracking. I told you in May that doubling the glass capacity was nowhere near enough. That is still true. Demand for glass is running ahead of supply even after the acquisition. So we are adding melting capacity in Orlando and Texas.

大口徑能力將開啟長距離相機系統、大型組件，以及太空基飛彈偵測與追蹤等應用。我在 5 月曾提到，玻璃產能加倍遠遠不夠。現在仍然如此。即使在收購之後，玻璃需求仍然超過供給。因此我們正在奧蘭多與德州新增熔融產能。

We are moving the AML, the Amorphous operation, into a larger building near our Visimid camera business in the Dallas area, and we're expanding downstream capacity in optical fabrication, coating, and assembly across the US and Latvian sites, including adding shifts in all locations. Al will talk about what that means for capital spending.

我們正將 AML（Amorphous 的營運）搬遷至達拉斯地區、靠近我們 Visimid 相機事業的更大建築，並在美國與拉脫維亞據點擴充下游產能，包括光學加工、鍍膜與組裝，同時在所有地點增加班次。Al 將說明這對資本支出意味著什麼。

The short version is that fiscal 2027 CapEx will be higher than fiscal 2026, and that is a deliberate choice made against a visible order book and pipeline.

簡而言之，2027 會計年度的資本支出（CapEx）將高於 2026 會計年度；這是在可見的訂單簿與管線支持下，我們刻意做出的選擇。

Before I hand it to Al, I'd like to discuss the balance sheet. In June, we raised $50 million in a primary offering alongside a secondary sale by North Run Capital, who funded our acquisition of G5.

在我把時間交給 Al 之前，我想先談談資產負債表。6 月，我們透過一項主要發行（primary offering）募集了 5,000 萬美元，同時 North Run Capital 也進行了次級出售（secondary sale）；North Run Capital 曾為我們收購 G5 提供資金。

We ended the fiscal year with $93.2 million of cash and effectively no debt. Some of that capital will be used to fund the capacity and working capital required to convert our growing backlog that has now grown over five consecutive quarters and continues to grow, and some will be to pursue accretive capability-adding acquisitions, such as the ones similar to G5 and AML that have proven we can execute and integrate.

我們在本財政年度結束時持有9,320萬美元現金，且實質上沒有負債。其中一部分資金將用於支應產能與營運資金需求，以轉化我們持續成長的在手訂單；該在手訂單已連續五個季度成長且仍在增加；另一部分則將用於追求具增益性的、可補強能力的併購，例如類似G5與AML的交易，這些案例已證明我們具備執行與整合的能力。

With that now, I will turn the call over to Al Miranda, CFO, to walk through the fourth quarter and full fiscal year 2026 financial results. Al, go ahead.

接下來，我將把電話會議交給財務長Al Miranda，請他帶大家回顧2026財年第四季與全年財務結果。Al，請開始。

Thank you, Sam. As always, I'll keep my review to a succinct highlight of the financials. Much of what we're discussing was also included in our press release issued earlier today, and will be included in the 10-K for the period. I encourage you to visit our Investor Relations webpage to access both documents.

謝謝你，Sam。一如往常，我會以精簡方式重點說明財務表現。我們今天稍早發布的新聞稿中也包含了我們討論的大部分內容，並將納入該期間的10-K報告。我鼓勵各位造訪我們的投資人關係網頁以取得這兩份文件。

In the fourth quarter, revenue increased 73% to $21.2 million, as compared to $12.2 million the same year ago quarter. Sales of infrared components were $7.1 million, 34% consolidated revenue. Visible components were $4.2 million, or 20% of consolidated revenue. Assemblies and modules were $9.1 million, or 43% of consolidated revenue. Engineering services were $0.8 million, or 4% of consolidated revenue.

第四季營收年增73%至2,120萬美元，相較去年同期的1,220萬美元。紅外線元件銷售為710萬美元，占合併營收34%。可見光元件為420萬美元，占合併營收20%。組件與模組為910萬美元，占合併營收43%。工程服務為80萬美元，占合併營收4%。

Gross profit increased 210% to $8.3 million, or 39.4% of revenue in the fourth quarter, as compared to $2.7 million, or 22% of revenue in the same year ago quarter. Sam mentioned the reasons for our gross margin increase, in addition to better absorption on higher production volume. And we also had a quarter carry through a $0.5 million of inventory reserve charge last year that didn't recur this year.

第四季毛利增加210%至830萬美元，占營收39.4%，相較去年同期為270萬美元、占營收22%。Sam已提到毛利率提升的原因，除了在較高產量下帶來更佳的成本吸收。此外，去年我們在該季度認列了50萬美元的存貨備抵費用並延續影響，但今年未再發生。

Operating expenses for the fourth quarter of fiscal 2026 were $12.6 million as compared to $7.2 million in the prior year period. Of the $5.4 million increase, $2 million relates to non-cash fair value adjustment to the acquisition earn-out liabilities, which are remeasured through operating expenses until fully settled. The increase is primarily related to G5 Infrared, reflecting its strong performance against the earn-out targets.

2026財年第四季營業費用為1,260萬美元，相較前一年同期為720萬美元。增加的540萬美元中，有200萬美元與併購或有對價（earn-out）負債的非現金公允價值調整有關；在完全結清前，該負債會持續透過營業費用重新衡量。此增加主要與G5 Infrared有關，反映其相對於earn-out目標的強勁表現。

The final earn-out amount was agreed to and accrued in the fourth quarter of fiscal 2026 to be paid in January 2027. The remaining operating expense increase of $3.4 million is primarily comprised of increased selling, general and administrative expenses, where the fourth quarter of fiscal 2026 included the addition of AML operations, incentive compensation accruals, additions to senior leadership team, higher sales and marketing investments, and continued information technology spend to meet customer security requirements.

最終的earn-out金額已於2026財年第四季達成協議並入帳提列，預計於2027年1月支付。其餘340萬美元的營業費用增加，主要來自銷售、一般及行政費用上升；其中2026財年第四季包含新增AML營運、獎酬性薪酬提列、擴充高階領導團隊、提高銷售與行銷投資，以及為符合客戶資安要求而持續投入資訊科技支出。

Net loss for the fourth quarter was $4.1 million, or $0.06 per basic and diluted share, compared to a net loss of $7.1 million, or $0.16 per share in the same year ago quarter. Adjusted EBITDA for the fourth quarter was $2.1 million, or 10% of revenue, compared to an adjusted EBITDA loss of $2 million in the year ago quarter. This is our fourth consecutive quarter of positive adjusted EBITDA.

第四季淨損為410萬美元，基本及稀釋後每股淨損0.06美元；相較去年同期淨損710萬美元、每股淨損0.16美元。第四季調整後EBITDA為210萬美元，占營收10%；相較去年同期調整後EBITDA為虧損200萬美元。這是我們連續第四個季度實現正的調整後EBITDA。

As I've said before, adjusted EBITDA is non-GAAP and not a perfect measure, but it is the better indicator of core operating performance because it strips out the non-cash acquisition accounting that otherwise dominates our reported results.

如我先前所說，調整後EBITDA屬於非GAAP指標，並非完美衡量方式，但它更能反映核心營運表現，因為它剔除了非現金的併購會計項目；否則這些項目會主導我們所報告的結果。

For the fiscal year, revenue for 2026 increased 93% to $71.7 million as compared to $37.2 million in the fiscal 2025. Sales of infrared components were $21.2 million, or 30%, an increase of 52% year over year. Visible components were $15.5 million, or 22% of consolidated revenue, an increase of 32%. Assemblies and modules were $31.9 million, or 44% of consolidated revenue, an increase of 281%. Engineering services were $3.1 million, or 4% of consolidated revenue, roughly flat with the prior year.

就全年而言，2026財年營收年增93%至7,170萬美元，相較2025財年為3,720萬美元。紅外線元件銷售為2,120萬美元，占30%，年增52%。可見光元件為1,550萬美元，占合併營收22%，年增32%。組件與模組為3,190萬美元，占合併營收44%，年增281%。工程服務為310萬美元，占合併營收4%，與前一年大致持平。

Operating expenses for fiscal 2026 were $45.5 million as compared to $22 million in the prior year. Of the $23.5 million increase, $14.1 million relates to the non-cash fair value adjustment to acquisition earn-out liabilities, which I will discuss further in a minute.

2026財年營業費用為4,550萬美元，相較前一年為2,200萬美元。增加的2,350萬美元中，有1,410萬美元與併購或有對價（earn-out）負債的非現金公允價值調整有關，我稍後會再進一步說明。

The remaining operating expense increase of $9.4 million reflects a full year of G5 Infrared operating cost, the addition of AML operating cost, higher sales and marketing spend, information technology investments to meet heightened customer security standards, and increased personnel cost associated with filling executive roles and accruing for incentive compensation plans.

其餘940萬美元的營業費用增加，反映G5 Infrared全年營運成本、加入AML的營運成本、更高的銷售與行銷支出、為符合更嚴格的客戶資安標準而進行的資訊科技投資，以及因補足高階主管職位與提列獎酬性薪酬計畫而增加的人事成本。

In addition, new product development costs also increased, which management views as an important part of execution of our strategy and plans to continue to grow our investments in new product developments. I want to be direct about the earn-out accounting because it is the largest single line in our income statement this year and the least reflective of our operating performance.

此外，新產品開發成本亦有所增加；管理層認為這是執行我們策略的重要一環，並計畫持續加大對新產品開發的投資。我想就earn-out會計處理直言不諱，因為它是今年損益表中最大的一個單一項目，也最不能反映我們的營運表現。

The $15.6 million charge in fiscal year '26 is a remeasurement of what G5 sellers earned. And it moved because G5 is outperforming the amount estimated at the time of acquisition, which per GAAP, was largely weighted based on historical financials. It is not an ongoing operating cost, and the majority of it is behind us, as we've now accrued for the final G5 earn-out to be paid in January 2027.

2026財年的1,560萬美元費用，是對G5賣方所賺取earn-out的重新衡量。之所以變動，是因為G5的表現優於併購時的估計；依GAAP規定，當時的估計在很大程度上是以歷史財務數據加權。這並非持續性的營運成本，而且其中大部分影響已經過去，因為我們已就最終的G5 earn-out完成提列，預計於2027年1月支付。

There may be small adjustments in future quarters related to AML and Visimid still to come.

未來季度可能仍會有與AML及Visimid相關的小幅調整。

Adjusted EBITDA for fiscal 2026 was a profit of $4.2 million compared to a loss of $5.1 million in fiscal 2025. As Sam said, a swing of $9.3 million. A good indicator of where we're heading. Backlog at June 30 was $110.9 million, up 197% from $37.4 million a year ago. Approximately $85.6 million of that is requested by customers for delivery within 12 months.

2026財年調整後EBITDA為獲利420萬美元，相較2025財年為虧損510萬美元。如Sam所說，改善幅度為930萬美元。這是我們發展方向的一個良好指標。截至6月30日，在手訂單為1億1,090萬美元，較一年前的3,740萬美元增加197%。其中約8,560萬美元為客戶要求於12個月內交付。

And then I'll touch on CapEx, as Sam mentioned. It is an indicator of how we are managing growth. In fiscal year '26, we spent $6.3 million in CapEx, $4.4 million in Q4 alone. The last time we spent at that level was in calendar year 2023 when we expanded the Orlando facility and doubled the size of the manufacturing and the clean room.

接著我會談一下資本支出（CapEx），如 Sam 所提到的。這是衡量我們如何管理成長的一項指標。在 26 財年，我們的 CapEx 為 630 萬美元，其中僅第四季就花了 440 萬美元。我們上一次在這個水準支出是在 2023 曆年，當時我們擴建了奧蘭多廠區，並將製造與無塵室的規模擴大一倍。

The fiscal '27 plan is larger and more ambitious. We'll expand all of our locations to meet the backlog we have and the demand we foresee through fiscal '27 and into fiscal year '28. The plan is to get ahead of demand in some key areas like the production of BlackDiamond glass.

27 財年的計畫規模更大、也更具企圖心。我們將擴充所有據點，以因應目前的積壓訂單，以及我們預期在 27 財年並延伸至 28 財年的需求。該計畫的目標是在一些關鍵領域提前布局需求，例如 BlackDiamond 玻璃的生產。

With that in mind, let me close with the frame I would use if I were on your side of the call. Two years ago, this was a $32 million per year revenue business with negative adjusted EBITDA, $3.5 million of cash. This fiscal year, it is a $72 million revenue business with positive adjusted EBITDA, $93 million cash, no debt, and $111 million order book that continues to grow.

基於此，讓我用一個框架作結：如果我站在各位這一端來聽這通電話，我會怎麼看。兩年前，這是一個年營收 3,200 萬美元、調整後 EBITDA 為負、現金 350 萬美元的業務。而在本財年，這是一個營收 7,200 萬美元、調整後 EBITDA 為正、現金 9,300 萬美元、零負債，且訂單簿達 1.11 億美元並持續成長的業務。

The work in fiscal 2027 is to add capacity fast enough, hold the margin gains, and convert the book. That is a straightforward execution mandate and it is fully funded.

27 財年的工作重點是：以足夠快的速度增加產能、守住毛利率的改善成果，並將訂單簿轉化為營收。這是一項直接明確的執行任務，而且資金已完全到位。

With that, I'll turn the call back to Sam for some closing remarks.

接下來我把電話交回給 Sam，請他做一些結語。

Thank you, Al, and thank you to everyone for joining us today.

謝謝你，Al，也謝謝今天所有加入我們的各位。

Fiscal 2026 was the year the pieces came together. A full year of G5, the addition of AML, a fortified balance sheet, a Western-aligned manufacturing base, and a set of programs that are moving from qualification into production. The upcoming fiscal 2027 will be different kind of year. It is less about proving the thesis and more about scaling against it.

2026 財年是各個拼圖到位的一年。G5 的完整年度貢獻、AML 的加入、更加穩健的資產負債表、與西方對齊的製造基地，以及一系列正從認證（qualification）走向量產的專案。即將到來的 2027 財年會是不同的一年。它較少著重於證明這個論點，而更著重於依此進行規模化擴張。

Doubling a manufacturing business in 12 months is hard. Doing it twice in a row is harder, and it will require capacity we are still building and people we are still hiring. We are clear-eyed about it, but we are doing it with $93 million of cash, no debt, two domestic glass plants, the broadest portfolio of infrared materials available, and a customer base that increasingly cannot source what they -- for what they make from anyone else in acceptable terms.

在 12 個月內讓一家製造業務規模翻倍很難。連續兩年做到就更難，而且需要我們仍在建置的產能，以及我們仍在招募的人才。我們對此有清醒的認知，但我們是在擁有 9,300 萬美元現金、零負債、兩座國內玻璃工廠、可取得的最廣泛紅外材料產品組合，以及一個愈來愈無法在可接受條件下從其他供應商取得其所需材料的客戶基礎之下來推進。

With that in mind, I want to close before we open for Q&A, by thanking the LightPath team. You, my team, delivered a record year while integrating two acquisitions, migrating infrastructure, and building capacity underneath a backlog that tripled. That is a great deal to ask of an organization our size, and you delivered it.

基於此，在我們進入問答之前，我想先感謝 LightPath 團隊。各位——我的團隊——在整合兩項併購、遷移基礎設施、並在積壓訂單增加到三倍的情況下建置產能的同時，仍交出創紀錄的一年。以我們這樣規模的組織而言，這是非常高的要求，而你們做到了。

With that in mind, I'll turn it back to the operator for Q&A. Operator?

接下來我把電話交回給主持人進行問答。主持人？

(Operator Instructions) Clarke Jeffries, Piper Sandler.

（主持人指示）Clarke Jeffries，Piper Sandler。

First question, just looking out to 2027, that $86 million that's sort of primed for delivery in 2027. Wondering if we could get a sense of the major programs that are embedded there and those two large contracts in July, are they set for 2027 delivery? And one follow-up.

第一個問題，著眼 2027 年，那 8,600 萬美元看起來是已經準備好在 2027 年交付的。想請問能否分享一下其中包含的主要專案，以及 7 月的那兩筆大型合約是否確定在 2027 年交付？另外還有一個追問。

Sure. So I'll answer the second part first. Yes, they're for 2027 delivery, the two large contracts that were in July. What's also nice about them or really encouraging about them, they're already production and both of them are a cadence of deliveries spread over months. And therefore we expect them to be renewed in the years to come as kind of programs of record typically do.

當然。我先回答第二部分。是的，7 月的那兩筆大型合約是 2027 年交付。另外令人欣慰、也很鼓舞的是，它們已經在量產中，而且兩者都是按節奏、分散在數個月內交付。因此我們預期它們在未來幾年會續約，因為這類「既定專案」（programs of record）通常就是如此。

Secondly, in terms of the breakout of it, Al, I don't know if you have it by product group, but application-wise, counter-UAS is still probably one of the biggest parts in there. Optics for drones and assemblies for drone dominance is a growing part in there. We don't have a lot of border patrol in there yet because the funding that has flown through to the primes has not come its way to us. So I'd say primarily counter-UAS right now is a very big driver.

第二，至於拆分，Al，我不確定你是否有按產品群組的資料；但從應用面來看，反無人機（counter-UAS）可能仍是其中最大的一塊。用於無人機的光學元件，以及用於無人機制空（drone dominance）的組件，在其中占比正在成長。我們目前在邊境巡防方面的占比還不高，因為流向主承包商（primes）的資金尚未傳導到我們這邊。所以我會說，目前主要驅動力仍是反無人機。

Perfect. And then just wondering if, Sam, you could give a little bit of additional context for G5 and the redesign time line. Just any reasons for the delay and maybe just a right setting on the earliest contribution you'd expect from those redesigned products.

了解。接著想請問 Sam，能否就 G5 與重新設計的時間線提供更多背景？包括延遲的原因，以及你們預期這些重新設計產品最早何時開始貢獻。

I think supply chains are stretched everywhere. And the redesign, while it doesn't require redesigning the entire camera, there's still mechanics and some other lenses that need to be redesigned, and some changes. So we're starting to feel the supply chain, especially on the detector side, impacting us across the board. So it really impacted that part, is one.

我認為供應鏈到處都很吃緊。而重新設計雖然不需要把整台相機都重新設計，但仍有機構件以及一些其他鏡頭需要重新設計，並且要做一些變更。因此我們開始感受到供應鏈的影響，尤其是在探測器端，對我們整體都造成影響。所以第一個影響點就在那一部分。

The second is, while Amorphous are able to melt the glass in that size, it is the first time or one of the first times Amorphous was melting that glass. And melting those kind of glass is not a plug-and-play recipe. I mean, even for us, from the moment we licensed the novel materials, it took us three years until we started producing them.

第二點是，雖然 Amorphous 能夠熔製那種尺寸的玻璃，但這是 Amorphous 第一次或最早幾次熔製那種玻璃。而熔製這類玻璃並不是把配方套上去就能直接運作（plug-and-play）。我的意思是，即便對我們而言，從我們取得新材料授權開始，也花了三年才開始生產。

So there's quite a bit of fine-tuning that once you start melting it. We're lucky to have an incredible team as we do at Amorphous that is able to do that at a really record time. But there's still more effort than just saying, okay, we'll melt it, and that's it.

所以一旦開始熔製，還需要相當多的微調。我們很幸運，Amorphous 擁有一支非常出色的團隊，能在非常破紀錄的時間內做到這點。但仍需要投入更多努力，而不只是說「好，我們把它熔了，就結束了」。

(Operator Instructions) Austin Moeller, Canaccord.

（主持人指示）Austin Moeller，Canaccord。

So just my first question. On the border patrol cameras or the equivalent overseas, are the overseas camera opportunities, are those foreign military sales to Middle East allies? Or is that the US Air Force and the Army deploying those on bases? And what could the TAM look like compared to border patrol?

我先問第一個問題。關於邊境巡防相機或海外的同等系統，海外的相機機會是對中東盟友的對外軍售（FMS）嗎？還是美國空軍與陸軍把這些部署在基地上？相較於邊境巡防，這部分的 TAM（可服務市場總量）可能會是什麼樣子？

Those are mostly foreign military sales to allies. The Air Force and so on, those are mainly counter-UAS systems, the different C-UAS systems. So I was talking about border patrol and towers along the borders of different Middle East or North African countries.

那些多半是對盟友的對外軍售。至於空軍等等，那些主要是反無人機系統，也就是不同的 C-UAS 系統。所以我剛才談的是邊境巡防，以及沿著中東或北非不同國家邊境設置的塔台。

The TAM is a bit difficult to tell because I don't know that market well enough. We're in it, sort of, we're one down. We're sub-primes there. But I would say that every deal like that, that comes in usually is in the tens of millions of dollars for us.

TAM 有點難判斷，因為我對那個市場不夠熟悉。我們算是有參與其中，但層級比較靠後。在那邊我們屬於次級（sub-prime）。但我會說，像那樣的每一筆案子，通常對我們而言規模都是數千萬美元。

Okay. And I understand you've shipped SPEIR cameras, at least initial prototypes to the Navy, and they installed them immediately. How many more SPEIR cameras can be made available to the Navy in the next 12 months and installed remotely in Southeast Asia, just given this tanker war, they're getting pelted by drones every day?

好的。我了解你們已經把 SPEIR 相機（至少是初期原型）交付給海軍，而且他們立刻就安裝了。考量到這場油輪戰爭、他們每天都遭到無人機襲擊，在未來 12 個月內，海軍還能取得並在東南亞以遠端方式安裝多少台 SPEIR 相機？

Well, I can't speak for the Navy and about their installation process or timelines. We did ship, I believe, the first camera, or maybe two, and I think we have another five cameras in dispatch before we get another order. And those five are expected to ship really in the next few weeks.

嗯，我無法代表海軍發言，也無法談他們的安裝流程或時程。我們確實出貨了，我相信是第一台相機，或可能兩台；而且我想在我們收到下一張訂單之前，還有另外五台相機正在準備出貨。這五台預期會在接下來幾週內出貨。

But beyond that, I really have no visibility into the process once we ship them until they make it onto the ships.

但除此之外，一旦我們出貨後，直到它們真正上艦之前，我對後續流程其實沒有能見度。

Richard Shannon, Craig-Hallum.

Richard Shannon，Craig-Hallum。

Let me ask a couple questions. Apologies for any ambient noise here in a loud area. Sam, I want to follow up on NGSRI. You made some prepared remarks here about not seeing any risk in the contracting proposal, I think is the term used by the government to the Army for this. I'd love to get a little more detail as to why you don't think that's a risk here.

我想問幾個問題。如果這裡環境很吵有背景噪音，先跟各位致歉。Sam，我想追問一下 NGSRI。你在事先準備的談話中提到，你沒有看到這份合約提案（我想這是政府對陸軍使用的用語）有任何風險。我希望你能多說一點，為什麼你認為這裡沒有風險。

I think that anyone who read this as I do the first time, it made it sound like they were dissatisfied in some manner with the current options here. And you've expressed a lot of confidence in the solution you're supporting there. So, would love to get a sense of why you think that's very low risk here, please.

我想任何人第一次像我這樣讀到它時，都會覺得它聽起來像是他們對目前的選項在某種程度上不滿意。而你也對你們所支持的解決方案表達了很高的信心。所以想請你說明一下，為什麼你認為這裡的風險非常低，謝謝。

Yes. So I'll refer here to an article that was published earlier this week. I can't remember the name of it. But the person there described, I think, pretty accurately that there are tens of thousands of launchers deployed in the field of the old Stinger missile. These new missiles need to be reverse compatible with those launchers. I'm not going to go into details of what the effort it takes to do that and what either Lockheed or Raytheon had to go through to do it.

是的。我這裡會引用本週稍早刊出的一篇文章。我記不起文章名稱了。但文中作者相當準確地描述：舊型刺針（Stinger）飛彈的發射器在前線部署了數以萬計。這些新飛彈必須能與那些發射器向後相容。我不會深入說明要做到這點需要多少努力，以及洛克希德或雷神為此必須經歷哪些流程。

I will point out that this program has been in the works for three years already, or even more. So this isn't something where even if you have a great missile, you can show up with it tomorrow morning and it will meet all the requirements and get in there. I view this more like -- to take an analogy, the XM30 or MX30 tanks, sorry, I forget which order of the letters are. Not tanks, sorry, carrier.

我要指出的是，這個計畫已經推動三年，甚至更久了。所以這不是那種就算你有一枚很棒的飛彈，明天早上拿來就能符合所有要求、立刻進入專案的情況。我把它看得更像——打個比方，XM30 或 MX30 戰車，抱歉，我忘了字母順序是哪個。不是戰車，抱歉，是運兵載具。

That just a couple of weeks ago, the Army said, okay, there are two main primes on this, same thing exactly, right? And we're going to open it up to see if by any chance anyone has someone else, something else to do our diligence properly. I think that the military organization as a whole, it seems that some of the neo-primes and new primes go off and often develop some solutions on their own dime, as they sort of, if we build it, they will come.

就在幾週前，陸軍說，好，這個案子有兩個主要主承包商（prime），完全一樣的情況，對吧？而我們要把它再打開，看看是否碰巧有人——其他人——有別的方案，讓我們能把盡職調查（diligence）做完整。我認為整體軍方組織似乎是這樣：一些新興主承包商和新的主承包商會離開去做，常常用自己的資金開發一些解決方案，抱著某種「只要我們做出來，他們就會來」的想法。

And it's doing their work properly, and I applaud them for that by saying, hey, before we stick to the same way we've been doing this for years, let's just check if someone happened to have developed something that fits the bill here and can do it. So I don't think this has come necessarily from being dissatisfied as much as it has come from the landscape is changing.

而這是在把工作做好；我也為此稱讚他們，因為他們說：嘿，在我們繼續沿用多年來同一套做法之前，先確認一下是否有人剛好開發出符合需求、能勝任的東西。所以我不認為這一定是出於不滿；更多是因為整個態勢正在改變。

If on all -- on most systems out there that you see, even if there are two primes, it gets opened up again to an industry day or call for RFIs, to see if there's something else that anyone else developed. I just think that it's highly unlikely that someone had something that was developed here and that can fit the bill exactly in this short time frame.

如果你看所有——或大多數——你看到的系統，即使已經有兩個主承包商，也常常會再度開放，辦產業日或徵詢 RFI，看看是否還有其他人開發了別的東西。我只是覺得，在這麼短的時間內，有人已經把東西開發到能完全符合需求的可能性非常低。

I mean, somebody would have to be well down the road in developing something right now if they were going to hit that April time frame next year, right? And I mean, the community for building these sorts of things is pretty small. I imagine this is probably well known if it exists.

我的意思是，如果有人要趕上明年四月那個時程，他現在就必須已經把開發推進到很後面的階段，對吧？而且，能做這類產品的社群其實很小。如果真有那樣的東西存在，我想大概早就眾所周知了。

So does that mean you don't think there's anyone out there that has that?

所以這代表你認為外面沒有人有那樣的東西嗎？

Well, I wouldn't quite say that. I don't know all the players out there and never say never, but I would say that both Lockheed and Raytheon are very, very experienced and really know what they're doing. And they've been working on this for three years. Everyone is welcome to reach their own conclusions. I don't want to jump to conclusions for other people.

嗯，我不會這麼斷言。我不認識所有參與者，也不會說絕不可能；但我會說，洛克希德和雷神都非常、非常有經驗，確實知道自己在做什麼。而且他們已經在這上面投入三年了。大家都可以自行得出結論。我不想替別人下結論。

My second question is in your prepared remarks, you also talked about some missile programs here, so I took good notes here about seven different platforms you're being evaluated for through with Lockheed Martin and some other ones.

我的第二個問題是，你在事先準備的談話中也提到一些飛彈計畫；我有做了很好的筆記，關於你們透過洛克希德馬丁以及其他一些案子，正在被評估用於七個不同的平台。

Maybe you could talk about some of these programs if you're able to, and maybe if you can convey either, average platform size or total TAM or something just to give us a sense here. Because it seems like missile programs can be pretty sizable like the NGSRI that you're already on. Maybe you can talk about the opportunity and kind of time frame to when those might be decided.

如果可以的話，你能談談其中一些計畫嗎？也許你能提供例如平均平台規模、或總 TAM 之類的資訊，讓我們有個概念。因為看起來飛彈計畫的規模可能相當可觀，就像你們已經參與的 NGSRI 一樣。也許你可以談談這個機會，以及大概什麼時程可能會做出決定。

So, yes, definitely. I have to be a bit careful here. First of all, some of them, we don't know enough about the end program to relate that. We know what we're told, but we don't know enough about it. Others might be very early or in a stage where they really don't want us talking about them.

所以，是的，當然可以。我這裡必須稍微謹慎一點。首先，有些案子我們對最終的計畫內容了解不夠，無法據此做連結或推估。我們知道的是對方告訴我們的，但我們了解得還不夠。另外一些可能還非常早期，或處在他們真的不希望我們談論的階段。

But let's just say that there was a talk by one of the generals recently from the Army talking specifically about air defense being a very layered approach, where you have short-range interceptors, mid-range, long-range, and even recently there was a call for space-based interceptors.

但我們就這麼說：最近陸軍有一位將軍發表談話，特別提到防空是一種非常分層的作法，你會有短程攔截器、中程、長程；甚至最近還有人呼籲要有天基攔截器。

So the Army and the Department of War as a whole need an entire range of products and not one. And there isn't one fits it all. So there is a very big push right now on building up capacity and refilling the inventory of THAAD and PAC-3, Patriot PAC-3. But there's probably an even bigger push at the next generation of all these interceptors.

因此，陸軍以及整個戰爭部門需要的是一整套產品，而不是單一產品。而且並不存在一個產品能夠一體適用。所以目前正大力推動擴充產能並補充 THAAD 與 PAC-3（愛國者 PAC-3）的庫存。但對於下一代所有這些攔截彈的推動力可能更大。

So I would say we're in almost all different types of interceptors or precision munition. When I say missiles, it's not only interceptors. Some of them are precision munition also. But I unfortunately don't have numbers really I can share at this point.

所以我會說，我們幾乎涵蓋各種類型的攔截彈或精準彈藥。我說的飛彈不僅僅是攔截彈。其中一些也是精準彈藥。但很遺憾，我目前沒有真正可以分享的數字。

Jon Hickman, Ladenburg Thalmann.

Jon Hickman，Ladenburg Thalmann。

I probably should know the answer to this, but could you explain to me what's a zoom camera?

我可能應該知道答案，但你能跟我解釋一下什麼是變焦相機嗎？

Yes. Zoom camera means it can change its field of view. So like in your iPhone, you can zoom in and out of the picture, right, magnify it, as opposed to a fixed focus camera. So cameras and optics we make for the small drones, those usually cannot zoom. They have a fixed field of view, you see one area because you're really trying to trade off weight and size. But the larger platforms all have the ability to zoom in and out.

可以。變焦相機是指它可以改變視場（field of view）。就像你的 iPhone，你可以把畫面放大或縮小，對吧，進行放大，而不是固定焦距的相機。我們為小型無人機製作的相機與光學元件，通常無法變焦。它們的視場是固定的，你只能看到一個區域，因為你其實是在重量與尺寸之間做取捨。但較大型的平台都具備放大與縮小的能力。

Okay. Okay. And then I have one more question. In the past, you've indicated that you thought you had a multi-year kind of runway before there might be other solutions for germanium to replace germanium other than your BlackDiamond glass. How do you feel about that now?

好的。好的。那我還有一個問題。過去你曾表示，你認為在出現其他可替代鍺（germanium）的解決方案之前，你大概還有一段多年的緩衝期，而不是只能靠你們的 BlackDiamond 玻璃。你現在對此的看法如何？

I feel like we actually have more time than I thought. I recently took a look at some of the announcements about germanium capacity. So you have on one hand, Umicore, the Belgium company that is considered the most promising in increasing capacity, and having a mine in Congo. They recently announced that what they're producing now is really only 3 to 4 tons of germanium a year.

我覺得我們實際上比我原先想的有更多時間。我最近看了一些關於鍺產能的公告。一方面，你看到 Umicore 這家比利時公司，被認為在提升產能方面最有希望，並且在剛果有礦場。他們最近宣布，目前的產量其實一年只有 3 到 4 噸鍺。

And what they expect to produce in a few years, and they're talking about four to five years, is -- it's more, but it's not definitely nowhere near solving all the problem. And then you look at some of the awards to companies in the US that are either on the mining or refining, and each one of them is talking about single digits tons of germanium. So I think we might have more time than I thought.

而他們預期在幾年後（他們談的是四到五年）能生產的量——確實會更多，但肯定遠遠不足以解決所有問題。再看看美國一些獲獎補助的公司，無論是採礦端或精煉端，每一家談的鍺產量也都是個位數噸。所以我認為我們可能比我原先想的有更多時間。

Okay. And then just so my math is correct, you reported a backlog of $110 million and then you reported these 2 big contracts for another $21 or $22 million. So if you add that to your backlog and then subtract whatever our estimates are for Q1, that should be close to a backlog for Q1?

好的。另外，為了確認我的計算正確，你們報告的積壓訂單（backlog）是 1.10 億美元，然後又公布了另外兩筆大合約，約 2,100 萬或 2,200 萬美元。所以如果把那加到你們的積壓訂單裡，再扣掉我們對第一季（Q1）的估計出貨金額，那應該就接近第一季的積壓訂單水位，對嗎？

Well, except that we also ship product out and we're continuing to grow. So I wouldn't quite say it's guaranteed that, that is a backlog. But it's definitely the backlog has been growing quite a bit since $110 million. So I think at the end of this quarter, we will have a very healthy backlog to report.

嗯，除非我們同時也在出貨，而且我們還在持續成長。所以我不會說那就一定保證是那樣的積壓訂單。但可以確定的是，自從 1.10 億美元以來，積壓訂單確實成長了不少。所以我認為在本季結束時，我們將能報告一個非常健康的積壓訂單。

Richard Shannon, Craig-Hallum.

Richard Shannon，Craig-Hallum。

Let me ask one more here. Sam, to follow up on the supply chain comments here, I'd love to get a sense of what are the kind of long poles in the tent here or the most difficult hurdles to overcome here for adding capacity across all the elements of what you're trying to do here. It's kind of holistically or qualitatively, and over what time frame do we expect to see this capacity increase completely?

我再問一個。Sam，延續你剛才對供應鏈的評論，我很想了解，在你們試圖在各個環節擴充產能的過程中，哪些是「帳篷裡最長的支柱」（long poles in the tent），也就是最難克服的障礙？從整體或定性角度來看，以及我們預期在什麼時間範圍內會看到這些產能完全提升？

Yes, that's a great question. And adding capacity in a vertically integrated company is kind of like a whack-a-mole game, if you would, because you solve one problem and it pops up elsewhere. However, right now, by far, glass is our biggest capacity constraint. We're adding a lot. That's where spending on CapEx is coming from. A lot of it is around the glass.

是的，這是個很好的問題。在一家垂直整合公司裡擴充產能，有點像打地鼠遊戲，因為你解決了一個問題，另一個地方又冒出來。不過，就目前而言，到目前為止，玻璃是我們最大的產能瓶頸。我們正在大量擴充。資本支出（CapEx）的投入主要就是來自這裡。其中很大一部分都與玻璃相關。

We will probably need to add even more. And we're adding it mainly for our systems, but also for some other customers or for subsystems. After the glass, the detectors are the long pole in the tent, which we don't have much control over. And the lead time on them has been growing from, I think, 6 months to 10 months, even more now.

我們可能還需要再增加更多。我們增加產能主要是為了我們的系統產品，但也有一部分是為了其他客戶或子系統。在玻璃之後，探測器（detectors）就是「帳篷裡最長的支柱」，而這部分我們能掌控的不多。它們的交期已從我記得的 6 個月拉長到 10 個月，現在甚至更久。

So you see that in our inventory where we are building up and preparing parts ahead of time and ordering them knowing that our vendors can't really react fast enough. After that, it really breaks down pretty quickly into many, many small things.

所以你會在我們的庫存中看到這點：我們正在提前備料、提前準備零件並下單，因為我們知道供應商實際上無法足夠快速地反應。在那之後，就會很快拆解成非常非常多的小項目。

Mechanics. Motors have become a really big deal because of the magnets in the motors now. It's the same pain everyone is feeling everywhere, I think, of making an integrated system.

機構件。馬達也變得非常關鍵，因為現在馬達裡的磁鐵問題。我想這是大家在各處都感受到的痛點：打造一個整合式系統。

Thank you. This concludes our question-and-answer session and today's event. Thank you for joining LightPath Technologies' fiscal fourth-quarter and full-year 2026 earnings conference call. You may now disconnect your lines.

謝謝。我們的問答環節以及今天的活動到此結束。感謝各位參加 LightPath Technologies 2026 會計年度第四季與全年財報電話會議。各位現在可以掛線。